Challenge 1 - Parametric Design by Computational Fluid Dynamics Simulation
DESCRIPTION
In the great majority of industrial problems involving fluid mechanics, turbulence effects must be taken into account. The physics of turbulence is extremely complex since non-linear effects lead to chaotic motion of the fluid, involving very different scales of space and time (multi-scale problem). In order to predict the fluid's mechanical behaviour, it is necessary to resolve the different scales as much as possible. Computational Fluid Dynamics (CFD) simulations of this type typically require considerable computational effort and time (in the order of weeks), for both designing the computational meshes and running the computer calculations.
Design optimization in the automotive industry is an interesting application of these type of numerical simulations; however, because they are currently unfeasible (due to their high computation cost), much faster prediction of the fluid flow is needed in order to run a design of experiments (DOE) with a combination of several parameters and multiple simulations.
OBJECTIVE
Implement effective techniques for accelerating the solution by:
• Identifying regions of the domain where the fluid flow characteristics are more influential for the problem, while saving computation cost by lowering the resolution in regions with lower turbulence intensity;
• Reducing the number of necessary numerical simulations with model reduction techniques (e.g. Proper Orthogonal Decomposition, Fast Fluid Dynamics, Scale-decoupling, etc.);
• Considering other effective techniques proposed by the participants.
A more detailed description of the application and state-of-the-art techniques will be given at the workshop.
Challenge 2 - Improvement of the Contact Center Performance
DESCRIPTION
Since 2013, The Eroski Group has been immersed in the SIEC (Integral, Efficient Customer Service) Project, the objectives of which are:• To provide a multipurpose contact point to internal customers.
• To define and implement the processes, technologies and equipment necessary to carry out that mission.
An important component of this project is the Contact Center, a specialized customer care team which carries out the tasks of contact with the customer, logging of incidents, correct escalation of incidents to specialist teams, and proper case closure.
OBJECTIVE
So that the Contact Center is efficient (i.e., it attends to customers with the appropriate measures according to the established service parameters), and taking into account that said contact center has a two-tiered structure (TIER 1 consisting of agents providing service in the first instance and TIER 2 teams composed of specialists in each area served), we seek a sizing model that allows us to properly manage the flow of incidents. Up to this point we have only been able to use (ERLANG) sizing methods for TIER 1, having found no model for TIER 2.
The proposed improvement project is directly related to optimizing resource allocation, reducing problem resolution time, and streamlining the process between the identification of a problem and the search for solutions. Some of the points and techniques that should be considered are:• Proposal of a model that describes the operation of the Contact Center service, along with the flow chart indicating the order of stages, priorities, etc.
• Review of the indicators that are currently used to catalogue incidents and their level of resolution, and definition of new measurable variables in the model that are useful for examining solutions to the problem.
• Analysis of the efficiency of processes, software and technologies currently used.
• A study of some aspects of the theory of Operational Research, particularly in network flow models, queuing theory or stochastic programming to formulate a deterministic version of the model that is capable of applying optimization methods.
Challenge 3 - Self-Organized Networks
DESCRIPTION
Fon is the world's leading carrier WiFi provider. Pioneers of residential WiFi sharing, we revolutionised carrier WiFi with our technology, creating a globally connected WiFi network. Today, we continue to innovate through two leading business areas. Fon Solutions offers best-in-class WiFi products and services. Our cutting-edge management solutions enable service providers to configure, deliver and operate their own WiFi services. Fon Network aggregates residential and premium carrier WiFi footprints creating one coherent global WiFi network. We facilitate WiFi interconnection between carriers, provide access deals to interested parties, and enable seamless user roaming. Fon´s global clients include British Telecom, the Deutsche Telekom Group, SFR, Proximus, KPN, Cosmote, MWEB, SoftBank, Telstra, and Vodafone.
Designed specifically with Communications Service Providers (CSPs) in mind, Fon’s cutting-edge WiFi Service Management Solution allows these companies to deliver WiFi services to subscribers and manage them just like cellular and fixed services, in a secure, scalable and flexible way.
WiFi networks are currently one of the main access technologies to the Internet, thanks to their low cost and easy deployment. However, their high density of WiFi access points may impact performance as the deployment is often unmanaged, unplanned, not coordinated in any way and consequently, far from optimal.
When a large number of WiFi hotspots are located within the same coverage area, it is likely that they operate in interfering frequencies with varying power levels. This has a severe impact on user performance due to the medium access mechanism defined in the 802.11 standard (CSMA/CA), whereby each user first listens to the medium and then only transmits if the listened channel is unoccupied.
OBJECTIVE
To develop an intelligent optimization algorithm to coordinate the frequency selection at the back end (for radio resource management purposes), in unmanaged, partially cooperative urban environments where not all the hotspots can be configured.
The expected outcome of the algorithm is to:➢ Minimize the number of interfering transmitters in the same contention domain, in areas where the spectrum is particularly crowded.
➢ Provide frequency channel planning at an urban district level.
Challenge 4 - Big Data in Sports: Predictive Models for Basketball Player's’ Performance
DESCRIPTION
Data analytics in professional sports has experienced rapid growth in recent years [1]. Development of predictive tools and techniques began to better measure both player and team performance. Statistics in basketball, for example, evaluate a player's and/or a team's performance [1,2].
Xpheres Basketball Management is one of the leading basketball player representation agencies in Spain and Europe. Xpheres has collected men’s professional basketball statistics from the last 16 seasons in more than 25 professional leagues and 71 FIBA tournaments. The complete database consists of more than 37,000 games and upwards of 20,000 players.
OBJECTIVE
Based on our historical database, we aim to:
1. Characterize the performance curve, peak and optimal age in professional men’s basketball using performance ratings of players in top European leagues.
2. Determine a rating correction factor for different basketball leagues, which accounts for intra-league and cross-league variability as well as for player characteristics (position, age, player ratings, etc.).
3. Determine which are the most important factors for predicting future outcomes (a successful professional career) of a basketball player.
4. Study statistical models to evaluate the performance of a player based on position, age, skills, league and other characteristics, and their influence in the game.
REFERENCES
[1] https://en.wikipedia.org/wiki/APBRmetrics#Common_statistics[2] https://en.wikipedia.org/wiki/Basketball_statistics
We believe that companies benefit a lot from the interactions between mathematicians and industrialists.
The open call to receive proposals closed on 23rd February 2017. However, if you think that your company may have an idea for a collaborative project, please do not hesitate to send an email to esgi131@bcamath.org